[Paranoia] Freshly damaged drive w/ cdparanoia?

[Bill Davidsen]

CJ Kucera wrote:
> Hello all.  Upon encounting a severely-scratched-up CD, I figured I'd
> let cdparanoia have a go at recovering the disc, so I started off a
> process to rip the tracks and ended up letting it go overnight because
> some of the tracks were taking ages to get through.
> 
> When I got back in the morning it had still been working on one of the
> tracks, and my xterm's scrollbuffer was all full of various errors 
> (unfortunately I didn't think to copy the errors into a text file or
> anything), and after awhile I ended up just killing the process,
> figuring that I'd clean off the disc again and try on the troublesome
> track again.
> 
> However, from that point on, it seems that the drive itself has become
> damaged somehow, and refuses to read *any* disc.  I'm curious if anyone's
> come across this behavior w/ cdparanoia before...  It seems to me that
> it's just got to be a cruel bit of coincidence that the drive should
> die in the middle of attempting to read a rather badly-damaged disc,
> but is there something in there which could actually cause some lasting
> damage on the drive?
> 
> Some details, and info on what I've done since then; if this is too far
> off-topic for the list, just ignore it:
> 
> The drive itself is actually a CD/DVD burner, Sony DW-U10A, being run
> in SCSI emulation mode.  I was running kernel 2.4.29 at the time of
> ripping, and I've successfully ripped many a disc with the drive (on
> the same day as these problems developed, no less).
> 
> Mechanically the drive seems fine.  It can eject and load discs with
> no problems, but when the disc goes in the drive, the amber light
> just stays on, the disc spins around (I know this because I've noted
> the orientation of the disc as I put it in, and it usually comes out
> in a different orientation) doing nothing else.  Eventually the amber
> light goes off, and no program I use is able to access the disc at all.
> 
> I tried rebooting, of course, and I tried several cold boots, once even
> unplugging the IDE cable inbetween, to no avail.  I tried booting into
> a 2.6.11 kernel to see if that would make a difference, which it didn't,
> and then I plugged the drive into another PC to see if it was actually
> the controller or maybe the IDE cable which was causing problems, which
> it wasn't (when on that PC, the drive would do the same thing; sit
> spinning for awhile when a disc was inserted, and nothing could access
> the drive).
> 
> So, I got to thinking that perhaps the firmware on the drive got FUBARed
> somewhere along the way.  The only actual update I could find for the
> DW-U10A was this crazy exe which was encoded primarily in Japanese
> which would have brought the firmware up to 1.1e, but it didn't seem to
> think that my drive was acceptable for the upgrade.  I *did* find a
> utility which claimed to softmod the drive into thinking that it was
> a DRU-500A (the DW-U10A is just the OEM version of said drive; apparently
> they're identical otherwise), and *that* patch applied without problems,
> though it didn't have any impact on drive behavior (apparently the drive
> *had* been at firmware 1.1d, for what it's worth).  After that I grabbed
> Sony's firmware update for the DRU-500A which would bring it up to 2.1a
> (from 1.0g on the other util), and that didn't have any impact on behavior
> either.
> 
> Now the drive's back in my main Linux box, and the drive is recognized
> properly by the kernel, a la:
> 
> 
>>scsi0 : SCSI host adapter emulation for IDE ATAPI devices
>>  Vendor: SONY      Model: DVD RW DRU-500A   Rev: 2.1a
>>  Type:   CD-ROM                             ANSI SCSI revision: 02
> 
> 
> ... and a scanbus in cdrecord seems to do all right with the drive, too:
> 
> 
>>  0,0,0     0) 'SONY    ' 'DVD RW DRU-500A ' '2.1a' Removable CD-ROM
> 
> 
> ... as you'd expect, but still no dice.
> 
> So yeah, if anyone's got any ideas, I'd appreciate hearing about 'em.
> Thanks, and apologies for the long-winded post!

Cleaning the drive seems like a logical next step at this point. If 
there was any dust on the damaged CD, or even if there weren't, it could 
be time for spring cleaning.

As an aside, I don't run paranoia on a burner, I have seen the 
suggestion that the head positioning mechanism on some drives is moving 
enough extra mass to accelerate wear and results in slower seeks. I 
*think* that's at least partially urban myth, but I have several 
machines which came with read-only drives which do a fair enough job and 
  cover the possibility that something could be damaged.

-- 
bill davidsen <davidsen at tmr.com>
   CTO TMR Associates, Inc
   Doing interesting things with small computers since 1979